40、协程

1、协程（greenlet模块）：在单线程中切换状态得模块  。greenlet模块封装在gevent内部。

import time
from greenlet import greenlet
def eat1():
    print('吃鸡腿1')
    g2.switch()          #线程切换到其他任务
    time.sleep(2)
    print('吃鸡腿2')
    g2.switch()


def eat2():
    print('吃饺子1')
    g1.switch()
    print('吃饺子2')


g1=greenlet(eat1)
g2=greenlet(eat2)
g1.switch()   

2、gevent：在实现切换线程得基础上又实现了规避IO　　

from gevent import monkey;monkey.patch_all()
import time     # time socket urllib requests
import gevent   # greenlet gevent在切换程序的基础上又实现了规避IO
from threading import current_thread
def func1():
    print(current_thread().name)
    print(123)
    time.sleep(1)
    print(456)

def func2():
    print(current_thread().name)   # dummythread
    print('hahaha')
    time.sleep(1)
    print('10jq')

g1 = gevent.spawn(func1)  # 遇见他认识的io会自动切换的模块
g2 = gevent.spawn(func2)
# g1.join()   #分别加入和joinall是一样的
# g2.join()
gevent.joinall([g1,g2])

3、爬取网页的例子

from gevent import monkey;monkey.patch_all()
import time
import gevent
import requests

# 爬取网页
# 10个网页
# 协程函数去发起10个网页的爬取任务
def get_url(url):
    res = requests.get(url)
    print(url,res.status_code,len(res.text))

url_lst =[
    'http://www.sohu.com',
    'http://www.baidu.com',
    'http://www.qq.com',
    'http://www.python.org',
    'http://www.cnblogs.com',
    'http://www.mi.com',
    'http://www.apache.org',
    'https://www.taobao.com',
    'http://www.360.com',
    'http://www.7daysinn.cn/'
]

start = time.time()
for url in url_lst:
    get_url(url)
print(time.time() - start)

from gevent import monkey;monkey.patch_all()
import time
import gevent
import requests

# 爬取网页
# 10个网页
# 协程函数去发起10个网页的爬取任务
def get_url(url):
    res = requests.get(url)
    print(url,res.status_code,len(res.text))

url_lst =[
    'http://www.sohu.com',
    'http://www.baidu.com',
    'http://www.qq.com',
    'http://www.python.org',
    'http://www.cnblogs.com',
    'http://www.mi.com',
    'http://www.apache.org',
    'https://www.taobao.com',
    'http://www.360.com',
    'http://www.7daysinn.cn/'
]

g_lst = []
start = time.time()
for url in url_lst:
    g = gevent.spawn(get_url,url)
    g_lst.append(g)
gevent.joinall(g_lst)
print(time.time() - start)